Fix unneeded stops and restarts in AD monitoring
When I execute "midclt call servicemonitor.stop", self.finished.set() is , which makes "self.finished.wait(self.frequency)" non-blocking; however, we don't stop the servicemonitor thread while "connected and started and enabled" are true.
This causes AD monitoring to go nuts and perform checks as quickly as possible, which in some environments causes the checks to fail and the AD service to either stop or restart. The latter option (restarting) usually succeeds, but causes a service interruption. Since the service monitor thread is still running, it will repeatedly restart causing fairly regular service disruptions.
Part of larger set of fixes to AD monitoring. PR#1340